Abstract
Recent studies established that some ferric salts, including FeCl3, decrease water permeability of the sweet cherry (Prunus avium L.) fruit exocarp and fruit cracking, presumably by a pH-dependent precipitation reaction that blocks high-flux pathways across the fruit surface. The objectives of our study were the following: to establish the effect of receiver pH on penetration of 55FeCl3 through excised exocarp segments (ES) and isolated cuticular membranes (CM) and to localize any Fe precipitates in the epidermal system of mature sweet cherry fruit. Penetration was studied using an infinite dose diffusion system where 55Fe penetrated from donor solutions of ferric salts (10 mm, pH 2.2–2.6) or EDTA-Na-Fe(III) (10 mm, pH 5.0) across an interfacing ES or CM into aqueous receiver solutions of pH values ranging from 2.0 to 6.0. For receiver pH 2.0, 55Fe penetration of the ES from a 10 mm FeCl3 donor (pH 2.6) was linear with time, but for receiver pH ≥ 3.0, penetration was low and insignificant. Increasing the pH of the water receiver from 2.0 to 6.0 in the course of an experiment resulted in an immediate halt of penetration regardless of whether 55Fe penetration occurred from FeCl3 (pH 2.6), Fe(NO3)3 (pH 2.6), or Fe2(SO4)3 (pH 2.4) as donor solutions (all at 10 mm). Only from EDTA-Na-Fe(III) (pH 5.0) 55Fe penetration continued to occur albeit at a decreased rate (−30%). At receiver pH 2.0, the 55FeCl3 flux through stomatous ‘Sam’ ES averaged 10.4 ± 2.3 pmol·m−2·s−1 and was positively correlated to stomatal density. Conventional and analytical electron microscopy (energy dispersive X-ray analysis, electron spectroscopic imaging, and electron energy loss spectroscopy) identified ferric precipitates in periclinal and anticlinal cell walls of epidermal cells underlying the cuticle, but not within the cuticle. These data indicate that the lack of 55Fe penetration from donor solutions of ferric salts through the ES into a receiver solution at pH ≥ 3 and the previously reported decrease in water uptake and cracking as a response to immersing fruit in solutions of ferric salts are the result of a precipitation reaction at the cuticle/cell wall interface in the sweet cherry exocarp. Although spray application of ferric salts is prohibitive for ecotoxicological reasons, understanding their mechanism in decreasing water uptake and fruit cracking may be helpful in the search for alternate compounds that are effective and ecotoxicologically acceptable.
Rain cracking of sweet cherry fruit is an important limitation in crop production worldwide and is thought to be related to water uptake through the fruit surface. This uptake occurs by diffusion through the cuticle and by viscous flow along an aqueous continuum across the sweet cherry fruit exocarp that is referred to as the polar pathway (Weichert and Knoche, 2006a). Polar pathways are formed by hydration and orientation of polar functional groups in the cuticular membrane [CM (Beyer et al., 2005; Schönherr and Bukovac, 1970; Schönherr and Schmidt, 1979; Schönherr, 2006)] and allow for rapid water uptake into the fruit (Weichert and Knoche, 2006a).
Incubating sweet cherry fruit in aqueous solutions of AgNO3, CuCl2, HgCl2, AlCl3, EuCl3, and FeCl3 was recently shown to markedly decrease water uptake and subsequent cracking (Beyer et al., 2002; Weichert et al., 2004). Unfortunately, none of these salts can be used in horticultural practice to control cracking. Nevertheless, understanding their mechanism in decreasing water uptake and fruit cracking may be helpful in identifying compounds that are equally effective as the above salts, but also acceptable from an ecotoxicological perspective.
Of the salts cited above, the most information is available on FeCl3. The data demonstrated that decreased water uptake is related to decreased permeability of the polar pathways (Weichert and Knoche, 2006b), which most likely results from a pH-dependent precipitation in the exocarp. Aqueous solutions of ferric salts [e.g. FeCl3, Fe(NO3)3, and Fe2(SO4)3] are strongly acidic. During penetration into the exocarp, ferric ions encounter a microenvironment of increasing pH until the pH of the fruit's apoplast is reached (Weichert and Knoche, 2006b). The increase in pH causes formation of colloidal gels and precipitates, comprising complex hydrated amorphous ferric-oxides and -hydroxides, most likely FeO(OH) (Greenwood and Earnshaw, 1990) that are expected to form a significant barrier to water transport across the sweet cherry exocarp (Weichert and Knoche, 2006b).
Indirect evidence for this pH-dependent precipitation reaction comes from experiments by Weichert and Knoche (2006b) where diffusion of 3H2O was studied using an infinite dose diffusion system (Bukovac and Petracek, 1993). In this system, 3H2O diffusion was followed from a donor solution containing FeCl3 (pH 2.6) through an interfacing ES into a water receiver of pH values adjusted to pH 2.0 to 6.0. In presence of FeCl3 in the donor, 3H2O diffusion through the ES into the receiver markedly decreased at receiver pH ≥ 3.0, but there was no effect of FeCl3 on 3H2O diffusion at the receiver pH 2.0 (Weichert and Knoche, 2006b). According to the above hypothesis, we would expect formation of precipitates at receiver pH ≥ 3.0 that must have decreased water uptake and fruit cracking.
To obtain direct evidence for this mechanism, we now studied the effect of receiver pH on penetration of 55Fe from ferric salts through excised ES or isolated CM into a receiver solution using the infinite dose diffusion system, and the occurrence of ferric or ferrous precipitates in the sweet cherry exocarp.
Materials and Methods
Plant material.
Mature sweet cherry (Prunus avium) cultivars Adriana, Sam, and Hedelfinger grafted on ‘Alkavo’ (P. avium) rootstocks were sampled from commercial orchards near Halle and experimental orchards of the Institute for Plant Breeding at Dresden-Pillnitz, and of the Lehr- and Versuchsanstalt Gartenbau at Erfurt, Germany. Fruit were selected for freedom from defects and uniformity of size and maturity based on color. The ES consisting of cuticle, epidermis, hypodermis, and some layers of mesocarp tissue were excised from the cheek region of ‘Adriana’ and ‘Sam’ sweet cherries using a razor blade. For the experiment on the effect of stomatal density (dsto), ES from ‘Sam’ fruit were also excised from stem cavity and the apical area of the fruit excluding the stylar scar to maximize the range in stomatal density (Peschel et al., 2003). ES were stored in 1 or 2 g·L−1 NaN3 at 5 °C until use. CM were isolated from the cheek of ‘Adriana’ sweet cherry fruit by incubating ES in pectinase and cellulase as described previously (Weichert and Knoche, 2006b).
Diffusion studies.
Diffusion of 55FeCl3 (specific activity 2.6 Gbq/mg, 99% radiochemical purity; PerkinElmer Life and Analytical Sciences, Boston) across sweet cherry ES and CM was determined using the infinite dose technique where diffusion is monitored under quasi steady-state conditions from a dilute donor through the interfacing ES or CM into a receiver (Bukovac and Petracek, 1993). ES or CM were mounted in Plexiglas holders (2 or 5 mm inner diameter) using silicone rubber (3140 RTV coating; Dow Corning, Midland, MI) and subsequently inspected for microscopic cracks in the cuticle at 100× magnification. ES or CM having microscopic cracks was discarded. Holders with crack-free ES or CM were mounted between two glass half-cells of the diffusion apparatus using silicone grease (Baysilone-Paste hochviskos; GE Bayer Silicones, Leverkusen, Germany). Generally, the morphological outer side of the ES faced the donor. Diffusion cells were placed in a thermostated water bath held at 25 °C and positioned on a multistirring plate. Unless specified otherwise, donor solutions were prepared 24 h before the initiation of an experiment using 10 mm FeCl3 (pH 2.6, FeCl3 × 6H2O; Carl Roth, Karlsruhe, Germany). Donor solutions were spiked with 55FeCl3 to yield concentrations of radioactivity ranging from 0.8 to 2.6 × 106 cpm/mL, equivalent to 0.07 to 0.24 μM 55FeCl3. Deionized water with pH adjusted using HCl or NaOH served as receiver solution because earlier studies established that some buffers interact with the Fe-dependent decrease in water transport (Weichert and Knoche, 2006b). The pH of receiver and donor solution was checked at the beginning and end of each experiment and remained within 0.1 pH units in all experiments. Diffusion experiments were initiated by adding 5 mL of donor and receiver solution to donor and receiver cells of diffusion units, respectively. The time course of penetration of 55Fe was followed by repeated sampling of the receiver solution. Aliquots (1 mL) were removed from receiver solutions, radioassayed by liquid scintillation spectrometry [scintillation cocktail (Ultima Gold™ XR; PerkinElmer Life and Analytical Sciences); counter (LS 6500; Beckman Instruments, Fullerton, CA)]; and replaced by fresh solution. Steady-state flow rates [F (mol·h−1)] were determined from the slope of a linear regression line fitted through plots of cumulative 55Fe penetration versus time. From F the flux [J (mol·m−2·s−1)] was calculated by dividing by the cross-sectional area of the ES or CM exposed in the diffusion cell [A (m2)].
The effect of receiver pH on the time course of 55Fe-transport from a 10 mm FeCl3 donor solution (pH 2.6) was studied using stomatous ES excised from ‘Sam’ sweet cherry fruit. Deionized water with pH adjusted to 2.0, 3.0, 4.5, and 6.0 using HCl or NaOH served as receiver. Because significant penetration was only detected at a receiver pH 2.0, the diffusion experiment was continued for the pH 2.0 treatment by increasing receiver pH to 6.0 at 47.5 h after initiation of the experiment. The minimum number of replication was four.
The effect of stomatal density on 55Fe-transport from a 10 mm FeCl3 donor solution (pH 2.6) was established in a two-phase experiment using astomatous ‘Adriana’ and stomatous ‘Sam’ ES. Stomata number was established by microscopy on an individual ES basis. ‘Adriana’ sweet cherry has a very low stomatal density (Peschel et al., 2003). When using holders having an inner diameter of the orifice of only 2 mm and ES or CM from ‘Adriana’ sweet cherry, it was possible to select astomatous specimen by microscopy. During phase I of the experiment [JI (‘Sam’, 0–48 h; ‘Adriana’, 0–96 h)], the pH of the water receiver was adjusted to 2.0. For phase II [JII (‘Sam’, 49–168 h)] the water receiver of pH 2.0 was replaced by a water receiver of pH 6.0 in ‘Sam’ ES. The astomatous ‘Adriana’ ES were discontinued because no detectable penetration occurred despite a 3-fold increase in concentration of radioactivity and extended sampling intervals. At the end of the experiment, the ES exposed in the holder was excised and transferred to a scintillation vial. After adding cocktail, radioactivity was quantified by liquid scintillation counting. Fluxes (J) were calculated from equilibrium flow rates as described above. The effect of pH on 55Fe-transport was indexed by the change in J [ΔJ (ΔJ = JI − JII)].
To investigate whether the decrease in 55Fe penetration from a 10 mm FeCl3 (pH 2.6) through stomatous Sam ES into deionized water (pH 5.5) depended on the orientation of the ES, penetration was studied through ES in normal (i.e., morphological outerside to the donor) versus reversed orientation (i.e., morphological outerside to the receiver). The J was established as described above.
The effect of the counter ion on the time course of 55Fe-transport was studied using donor solutions containing FeCl3 (FeCl3 × 6H2O; Carl Roth), Fe(NO3)3 [Fe(NO3)3 × 9H2O; Sigma-Aldrich, St. Louis], Fe2(SO4)3 [Fe2(SO4)3 × H2O; Sigma-Aldrich], or EDTA-Na-Fe(III) (Sigma-Aldrich; all at 10 mm). Donor solutions were spiked with 55FeCl3 24 h before initiation of the experiment to yield a concentration of radioactivity of about 0.8 × 106 cpm/mL. Steady-state fluxes were established at receiver pH 2.0 (JI, 0–47 h) and subsequently at receiver pH 6.0 (JII, 47.5–95.5 h). The effect of pH on 55Fe transport was indexed by the ratio JII/JI.
Electron microscopy and elemental analysis
Sample preparation.
Unless otherwise specified, ‘Hedelfinger’ fruit were selected for localizing Fe precipitates because this cultivar has a high frequency of stomata (Peschel et al., 2003). Fruit were incubated for 10 h in 10 mm FeCl3, removed from solution, thoroughly rinsed with deionized water, and carefully blotted using tissue paper. Small segments (4 mm2 area, thickness 1–2 mm) of exocarp and adhering mesocarp tissue were excised using a razor blade. Segments used for Transmission electron microscopy (TEM) were fixed with 3% glutaraldehyde in 0.6 M phosphate buffer (pH 7.4) and 1% OsO4 in 0.1 M Palade buffer [pH 7.0 (Palade, 1952)], and those used for analytical electron microscopy with 3% glutaraldehyde in 0.6 M phosphate buffer (pH 7.4) only. Subsequently, samples were dehydrated in an ascending series of acetone and embedded in low-viscosity epoxy resin [ERL (Plano, Wetzlar, Germany)].
TEM.
Ultrathin sections (50 nm) were investigated with a transmission electron microscope (model EM912 OMEGA; Leo, Oberkochen, Germany) equipped with an incolumn energy filter and an energy dispersive X-ray (EDX) analysis System (Link exIII; Oxford Instruments, High Wycombe, UK). The images were recorded with a 2K slow scan charged-coupled device (CCD)-camera (Proscan elektronische Systeme, Lagerlechfeld, Germany). EDX analysis was performed on exocarp cross sections of 150 to 200 nm thickness. Segments were collected on Cu-grids and analyzed in the spot mode (spot size 100 nm at 80 keV acceleration voltage, emission current 20 μA).
Electron spectroscopic imaging (ESI) and electron energy loss spectroscopy (EELS).
The spatial distribution of Fe and Cl across the exocarp and the valency of the Fe in the precipitates were established using ESI and EELS, respectively. Briefly, ultrathin sections (50 nm) on Cu-grids were investigated in an EM910 OMEGA electron microscope (Leo). Images and spectra were recorded and subsequently processed by digital image analysis (analySIS 3.1; Soft Imaging System, Münster, Germany). For localization of precipitates, Fe or Cl maps were superimposed with the corresponding zero loss images. Superimposing the Fe and Cl signal simultaneously with the zero loss image allowed the colocalization of Fe and Cl in the exocarp.
Electron energy loss spectra were recorded from the sweet cherry fruit exocarp and were compared with spectra of Fe2+ and Fe3+standards.
Statistics.
Data were subjected to analysis of variance (ANOVA). ANOVA (Proc Anova, Proc Glm), multiple comparisons of means, and linear regression analysis (Proc Reg) were carried out using SAS (version 9.1; SAS Institute, Cary, NC). Unless specified otherwise, regression analysis was performed on data for individual ES. Data in figures are presented as means ± se of means except for Fig. 2, A–C, where data points represent individual ES.
Results
Penetration of 55Fe from a FeCl3 donor across the sweet cherry exocarp increased linearly with time at the receiver pH 2.0 (Fig. 1A). There was essentially no penetration at receiver pH ≥ 3.0. Replacing the receiver of pH 2.0 by a receiver of pH 6.0 resulted in an immediate halt of penetration (Fig. 1B). The flux (J) calculated from mean flow rates of 55Fe across stomatous ‘Sam’ ES at a receiver pH 2.0 averaged 10.44 ± 2.34 pmol·m−2·s−1 at a donor concentration of 10 mm FeCl3 (n = 16 ES).
Effect of receiver pH on the time course of 55Fe-transport across exocarp segments (ES) excised from mature ‘Sam’ sweet cherry fruit. (A) Penetration at receiver pH 2.0, 3.0, 4.5, and 6.0. (B) Change in 55Fe-transport across ES upon increasing the receiver pH from 2.0 during phase I to 6.0 during phase II.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Effect of receiver pH on the time course of 55Fe-transport across exocarp segments (ES) excised from mature ‘Sam’ sweet cherry fruit. (A) Penetration at receiver pH 2.0, 3.0, 4.5, and 6.0. (B) Change in 55Fe-transport across ES upon increasing the receiver pH from 2.0 during phase I to 6.0 during phase II.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Effect of receiver pH on the time course of 55Fe-transport across exocarp segments (ES) excised from mature ‘Sam’ sweet cherry fruit. (A) Penetration at receiver pH 2.0, 3.0, 4.5, and 6.0. (B) Change in 55Fe-transport across ES upon increasing the receiver pH from 2.0 during phase I to 6.0 during phase II.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
55Fe penetration from a FeCl3 donor into a receiver of pH 2.0 was a linear function of stomatal density (Fig. 2A). The regression line for the relationship between J and dsto passed through the origin and predicted the absence of 55Fe penetration in an astomatous system (dsto = 0 mm−2). Indeed, for astomatous ‘Adriana’ CM and ES, the J were very low (2.01 ± 0.66 and 2.73 ± 1.12 pmol·m−2·s−1 for ‘Adriana’ CM and ES, respectively; both at receiver pH 2.0). Increasing receiver pH from 2.0 to 6.0 for stomatous ‘Sam’ ES resulted in a drastic decrease in J (Fig. 2B). This decrease was again a linear function of stomatal density (Fig. 2C). The y-axis intercept of the regression line represented the decrease in J for a hypothetical astomatous ES (dsto = 0 mm−2) and did not differ from zero. The decrease in J upon increasing the receiver pH was closely related to the amount of 55Fe associated with the ES at the end of the experiment (Fig. 2C, inset).
Effect of stomatal density (dsto) on the flux (J) of 55Fe through astomatous ‘Adriana’ and stomatous ‘Sam’ sweet cherry fruit exocarp segments (ES). The J for 55Fe-transport was determined (A) at receiver pH 2.0 during phase I (JI) and (B) at receiver pH 6.0 during phase II (JII). (C) Decrease in J for 55Fe-transport (ΔJ) upon increasing receiver pH from 2.0 to 6.0. Inset: ΔJ as affected by the amount of 55Fe in the ES. The ΔJ was calculated by subtracting the J at a receiver pH 6.0 (JII) from that at pH 2.0 (JI). Regression equations for the relationship between J and dsto were: receiver pH 2.0: JI (pmol·m−2·s−1) = 0.55 (±2.29) + 7.04 (±1.30) × dsto (mm−2), r2 = 0.68***, P ≤ 0.0001; receiver pH 6.0: JII (pmol·m−2·s−1) = −0.17 (±0.27) + 0.55 (±0.16) × dsto (mm−2), r2 = 0.47**, P ≤ 0.003. The corresponding equation for the decrease in ΔJ upon increasing receiver pH from 2.0 to 6.0 was: ΔJ (pmol·m−2·s−1) = 0.72 (±2.06) + 6.49 (±1.17) × dsto (mm−2), r2 = 0.69***, P ≤ 0.0001; that for the relationship between the decrease in ΔJ and the amount of 55Fe in the ES at the end of the experiment: ΔJ (pmol·m−2·s−1) = −7.83 (±2.97) + 1.35 (±0.21) × Amt 55Fe (pmol), r2 = 0.75***, P ≤ 0.0001. Data for ‘Sam’ represent individual ES, those for ‘Adriana’ mean ± se of eight astomatous ES.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Effect of stomatal density (dsto) on the flux (J) of 55Fe through astomatous ‘Adriana’ and stomatous ‘Sam’ sweet cherry fruit exocarp segments (ES). The J for 55Fe-transport was determined (A) at receiver pH 2.0 during phase I (JI) and (B) at receiver pH 6.0 during phase II (JII). (C) Decrease in J for 55Fe-transport (ΔJ) upon increasing receiver pH from 2.0 to 6.0. Inset: ΔJ as affected by the amount of 55Fe in the ES. The ΔJ was calculated by subtracting the J at a receiver pH 6.0 (JII) from that at pH 2.0 (JI). Regression equations for the relationship between J and dsto were: receiver pH 2.0: JI (pmol·m−2·s−1) = 0.55 (±2.29) + 7.04 (±1.30) × dsto (mm−2), r2 = 0.68***, P ≤ 0.0001; receiver pH 6.0: JII (pmol·m−2·s−1) = −0.17 (±0.27) + 0.55 (±0.16) × dsto (mm−2), r2 = 0.47**, P ≤ 0.003. The corresponding equation for the decrease in ΔJ upon increasing receiver pH from 2.0 to 6.0 was: ΔJ (pmol·m−2·s−1) = 0.72 (±2.06) + 6.49 (±1.17) × dsto (mm−2), r2 = 0.69***, P ≤ 0.0001; that for the relationship between the decrease in ΔJ and the amount of 55Fe in the ES at the end of the experiment: ΔJ (pmol·m−2·s−1) = −7.83 (±2.97) + 1.35 (±0.21) × Amt 55Fe (pmol), r2 = 0.75***, P ≤ 0.0001. Data for ‘Sam’ represent individual ES, those for ‘Adriana’ mean ± se of eight astomatous ES.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Effect of stomatal density (dsto) on the flux (J) of 55Fe through astomatous ‘Adriana’ and stomatous ‘Sam’ sweet cherry fruit exocarp segments (ES). The J for 55Fe-transport was determined (A) at receiver pH 2.0 during phase I (JI) and (B) at receiver pH 6.0 during phase II (JII). (C) Decrease in J for 55Fe-transport (ΔJ) upon increasing receiver pH from 2.0 to 6.0. Inset: ΔJ as affected by the amount of 55Fe in the ES. The ΔJ was calculated by subtracting the J at a receiver pH 6.0 (JII) from that at pH 2.0 (JI). Regression equations for the relationship between J and dsto were: receiver pH 2.0: JI (pmol·m−2·s−1) = 0.55 (±2.29) + 7.04 (±1.30) × dsto (mm−2), r2 = 0.68***, P ≤ 0.0001; receiver pH 6.0: JII (pmol·m−2·s−1) = −0.17 (±0.27) + 0.55 (±0.16) × dsto (mm−2), r2 = 0.47**, P ≤ 0.003. The corresponding equation for the decrease in ΔJ upon increasing receiver pH from 2.0 to 6.0 was: ΔJ (pmol·m−2·s−1) = 0.72 (±2.06) + 6.49 (±1.17) × dsto (mm−2), r2 = 0.69***, P ≤ 0.0001; that for the relationship between the decrease in ΔJ and the amount of 55Fe in the ES at the end of the experiment: ΔJ (pmol·m−2·s−1) = −7.83 (±2.97) + 1.35 (±0.21) × Amt 55Fe (pmol), r2 = 0.75***, P ≤ 0.0001. Data for ‘Sam’ represent individual ES, those for ‘Adriana’ mean ± se of eight astomatous ES.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
The effect of receiver pH was independent of the orientation of the ES relative to the 55FeCl3 donor. When water at pH 5.5 served as the receiver, the J for 55Fe transport from FeCl3 averaged 0.11 ± 0.03 and 0.09 ± 0.03 pmol·m−2·s−1 for normal versus reverse orientation of the ES, respectively.
There was no significant difference in J for 55Fe penetration from FeCl3, Fe(NO3)3, and Fe2(SO4)3, or EDTA-Na-Fe(III) at the receiver pH 2.0 (Fig. 3, Table 1). However, increasing receiver pH from 2.0 to 6.0 nearly stopped 55Fe penetration from the salts FeCl3, Fe(NO3)3, and Fe2(SO4)3. In contrast, 55Fe penetration from EDTA-Na-Fe(III) continued to increase, albeit at a decreased rate [−30% (Table 1)].
Effect of counter ion, chelator, and receiver pH on penetration of 55Fe through excised exocarp segments of ‘Sam’ sweet cherry fruit. During phase I of the experiment, permeability was established at a receiver pH 2.0 (JI), in phase II, at a receiver pH 6.0 (JII).
Effect of counterion, chelator, and receiver pH on the time course of 55Fe-transport across ‘Sam’ exocarp segments excised from mature sweet cherry fruit. During phase I, 55Fe-transport was determined at receiver pH 2.0, during phase II, at receiver pH 6.0. For flux estimates, see Table 1.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Effect of counterion, chelator, and receiver pH on the time course of 55Fe-transport across ‘Sam’ exocarp segments excised from mature sweet cherry fruit. During phase I, 55Fe-transport was determined at receiver pH 2.0, during phase II, at receiver pH 6.0. For flux estimates, see Table 1.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Effect of counterion, chelator, and receiver pH on the time course of 55Fe-transport across ‘Sam’ exocarp segments excised from mature sweet cherry fruit. During phase I, 55Fe-transport was determined at receiver pH 2.0, during phase II, at receiver pH 6.0. For flux estimates, see Table 1.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
TEM revealed a thin cuticle of about 1 μm thickness on cell walls that averaged about 3 μm in thickness in the periclinal region of epidermal cells (Fig. 4A). Micrographs from sections of fruit incubated in 10 mm FeCl3 showed electron dense particles in anticlinal and periclinal cell wall regions, particularly at the cuticle/cell wall interface (Fig. 4B). The frequency of these electron-dense particles decreased as distance from the cuticle increased (Fig. 4B). Using ESI and EDX, these particles were identified as Fe precipitates (Figs. 4, C–G and 5). There was no detectable Fe signal (red) in the cuticle (Fig. 4C). Cl maps (green) generated by ESI revealed the presence of Cl throughout the cuticle, the cuticle/cell wall interface, and the cell wall region (Fig. 4D). The overlay map of Cl (green) and Fe (red) indicated by the yellow color that some Fe may be associated with Cl (Fig. 4E). Comparing energy loss spectra of the Fe precipitates in the sweet cherry fruit exocarp with those of Fe2+ and Fe3+ standards demonstrated that precipitates contained ferric but not ferrous Fe (Fig. 6).
Cross sections and elemental distribution maps of Fe and Cl in the sweet cherry fruit exocarp. (A) Anticlinal and periclinal cell wall region of the epidermis of ‘Sam’ sweet cherry. Zero loss images (B and E) and elemental distribution maps of Fe [C (red)], Cl [D (green)] and the overlay image of Fe and Cl [E (yellow)] in periclinal and anticlinal regions (F and G) of the sweet cherry exocarp following incubation of ‘Hedelfinger’ fruit in 10 mm FeCl3. Elemental distribution maps were obtained by electron spectroscopic imaging (ESI). For details see materials and methods; CM = cuticular membrane, CW = cell wall.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Cross sections and elemental distribution maps of Fe and Cl in the sweet cherry fruit exocarp. (A) Anticlinal and periclinal cell wall region of the epidermis of ‘Sam’ sweet cherry. Zero loss images (B and E) and elemental distribution maps of Fe [C (red)], Cl [D (green)] and the overlay image of Fe and Cl [E (yellow)] in periclinal and anticlinal regions (F and G) of the sweet cherry exocarp following incubation of ‘Hedelfinger’ fruit in 10 mm FeCl3. Elemental distribution maps were obtained by electron spectroscopic imaging (ESI). For details see materials and methods; CM = cuticular membrane, CW = cell wall.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Cross sections and elemental distribution maps of Fe and Cl in the sweet cherry fruit exocarp. (A) Anticlinal and periclinal cell wall region of the epidermis of ‘Sam’ sweet cherry. Zero loss images (B and E) and elemental distribution maps of Fe [C (red)], Cl [D (green)] and the overlay image of Fe and Cl [E (yellow)] in periclinal and anticlinal regions (F and G) of the sweet cherry exocarp following incubation of ‘Hedelfinger’ fruit in 10 mm FeCl3. Elemental distribution maps were obtained by electron spectroscopic imaging (ESI). For details see materials and methods; CM = cuticular membrane, CW = cell wall.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Energy dispersive X-ray analysis (EDX) of cross sections of the sweet cherry fruit exocarp after incubation in 10 mm FeCl3. EDX spectra were obtained in the cuticle (A) and the epidermal cell wall region with Fe deposits (B). The Cu peaks were caused by the supporting grid.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Energy dispersive X-ray analysis (EDX) of cross sections of the sweet cherry fruit exocarp after incubation in 10 mm FeCl3. EDX spectra were obtained in the cuticle (A) and the epidermal cell wall region with Fe deposits (B). The Cu peaks were caused by the supporting grid.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Energy dispersive X-ray analysis (EDX) of cross sections of the sweet cherry fruit exocarp after incubation in 10 mm FeCl3. EDX spectra were obtained in the cuticle (A) and the epidermal cell wall region with Fe deposits (B). The Cu peaks were caused by the supporting grid.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Electron energy loss spectra (EELS) of Fe precipitates in the periclinal cell wall region of the epidermis of the sweet cherry fruit exocarp (solid line). The Fe2+ (dashed line) and Fe3+ signal (dotted line) originated from Fe(OH)2 and Fe(OH)3 that served as standards of known valency for comparison.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Electron energy loss spectra (EELS) of Fe precipitates in the periclinal cell wall region of the epidermis of the sweet cherry fruit exocarp (solid line). The Fe2+ (dashed line) and Fe3+ signal (dotted line) originated from Fe(OH)2 and Fe(OH)3 that served as standards of known valency for comparison.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Electron energy loss spectra (EELS) of Fe precipitates in the periclinal cell wall region of the epidermis of the sweet cherry fruit exocarp (solid line). The Fe2+ (dashed line) and Fe3+ signal (dotted line) originated from Fe(OH)2 and Fe(OH)3 that served as standards of known valency for comparison.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 135, 2; 10.21273/JASHS.135.2.95
Discussion
Our data on 55Fe penetration and the localization of ferric iron in the epidermal cell walls are consistent with the hypothesis of a pH-dependent precipitation reaction in the sweet cherry exocarp. There was little penetration of 55Fe from donor solutions of the salts FeCl3, Fe(NO3)3, and Fe2(SO4)3 into receiver solutions at pH values ranging from 3.0 to 6.0 because of formation of ferric precipitates in the exocarp. However, penetration was up to 36-fold higher at the receiver pH 2.0 where precipitates would not form. The immediate cessation of 55Fe penetration from FeCl3 into the receiver upon increasing the receiver pH from 2.0 to 6.0 demonstrates that the ES in the infinite dose system must have equilibrated rapidly with the pH of the receiver solution and the ferric precipitates formed instantaneously.
Support for a causal relationship between the formation of ferric precipitates as indexed by the absence of significant Fe penetration into the receiver and the effect of ferric salts on water uptake comes from a detailed comparison of data on 55Fe penetration (this study) and previously published data on 3H2O diffusion through ES, osmotic water uptake, and cracking in intact fruit (Beyer et al., 2002; Weichert et al., 2004; Weichert and Knoche, 2006b). First, receiver pH greater than or equal to 3.0 had a marked negative effect on penetration of 55Fe from FeCl3 (Fig. 1) and on water transport in presence of FeCl3 (Beyer et al., 2002; Weichert and Knoche, 2006b). Second, astomatous ES (and CM) were essentially impermeable to 55Fe from FeCl3 (Fig. 2) and we did not detect any effect of FeCl3 on water transport in astomatous systems (Weichert and Knoche, 2006b). Third, 55Fe penetration across the exocarp into the receiver at pH 2.0 (Fig. 2) and 3H2O penetration were positively related to stomatal density (Weichert and Knoche, 2006b). When increasing the pH of the receiver from 2.0 to 6.0, 55Fe penetration from the donor into the receiver markedly decreased. Similarly, 3H2O penetration into the receiver markedly decreased when adding FeCl3 to the donor (Weichert and Knoche, 2006b). In both experiments, the ferric ion in the donor must have penetrated into and precipitated within the exocarp, resulting in a marked decrease of 55Fe and 3H2O penetration through the exocarp into the receiver solution. Fourth, the marked decrease of 55Fe penetration upon increasing receiver pH from 2.0 to 6.0 (Fig. 1B) and the decrease in 3H2O penetration caused by FeCl3 (Weichert and Knoche, 2006b) were independent of the orientation of the ES. Fifth, FeCl3, Fe(NO3)3, and Fe2(SO4)3 decreased water permeability of stomatous ES and intact fruit (Beyer et al., 2002). Accordingly, there was little 55Fe penetration from these salts at a receiver pH 6.0, which mimics the pH of the cell wall space underlying the CM (Fig. 3; Marschner, 1995). In contrast, 55Fe penetration occurred from EDTA-Na-Fe(III) in the donor at receiver pH 6.0 and the chelate had no effect on water uptake into sweet cherry fruit (H. Weichert, unpublished data). Thus, experimental conditions that induced formation of ferric precipitates in the exocarp markedly decreased 55Fe and 3H2O penetration through the exocarp into the receiver and water uptake and cracking of detached sweet cherry fruit (Weichert et al., 2004). These three conditions were the following: 1) a stomatous exocarp, 2) free ferric ions in the donor, and 3) a receiving compartment (i.e., a receiver solution of the infinite dose system or the apoplast of an intact fruit) having a pH ≥ 3.0.
It may be argued that the ligand exchange between the radiolabeled 55FeCl3 used for spiking and the non-radiolabeled Fe of the EDTA-Na-Fe(III) complex may have been incomplete and that this would have resulted in penetration of 55FeCl3 rather than of EDTA-Na-55Fe. However, the following arguments indicate that this was unlikely the case in our study. First, for EDTA–Na-Fe(III) (stability constant K, log K = 25.7; Furia, 1972), half times of ligand exchange of 55Fe and EDTA–Na-Fe(III) depended on pH and decreased from 1480 to 740 min as pH increased from 1.6 to 2.5 (Jones and Long, 1952). Thus, at pH 2.5 two half-times would have past when donor solutions were prepared 24 h before the experiment, as was done in our study. In our study, pH of the EDTA–Na-Fe(III) donor solution was markedly higher and, therefore, half-times most likely were even shorter. Second, if ligand exchange was incomplete, the 55FeCl3 in the donor solution would have precipitated immediately in the donor as complex hydrated amorphous ferric-oxides and -hydroxides. Under these circumstances, 55Fe would not be detectable in the donor. Third, the data obtained for 55FeCl3 penetration demonstrate conclusively that even if 55FeCl3 had remained in the donor in solution, the high pH of the receiver would have caused precipitation in the ES and the absence of detectable penetration into the receiver solution (Fig. 1). However, this was not the case. Thus, incomplete ligand exchange was unlikely to be a factor in our study.
The close relationship between 55Fe penetration from FeCl3 and stomatal density or between the FeCl3 induced decrease in water permeability and stomatal density (Weichert and Knoche, 2006b) deserves some further comments. In sweet cherry fruit, as in many other species (Franke, 1964; Schlegel et al., 2005; Schönherr and Bukovac, 1970; Schönherr, 2006), polar pathways are primarily associated with stomata (Weichert and Knoche, 2006a). Because water transport across the exocarp occurs predominantly via these polar pathways (Weichert and Knoche, 2006a) and FeCl3 decreased water uptake (Weichert and Knoche, 2006b), penetration of 55Fe ions from ferric salt solutions must also occur along these pathways. This conclusion is supported by two further arguments. First, the ferric ion carries three charges and thus, is excluded from penetration along the lipophilic pathway (Schönherr et al., 2005). Second, comparison of the size exclusion limits of polar pathways [between 0.47 and 1.15 nm; estimated from reflection coefficients (σ) for sucrose (σ = 0.74) and polyethylene glycol 1500 (σ = 1.0; Weichert and Knoche, 2006a)] and the hydrodynamic radius of the ferric ion (0.41 nm; Beyer et al., 2002) revealed that ferric ions are accommodated by the polar pathway. Unfortunately, we were not successful in obtaining cross sections through a stoma to provide direct evidence for preferential precipitation at stomata. Although ‘Hedelfinger’ sweet cherry has a high number of stomata relative to fruit from other cultivars, stomatal density on sweet cherry fruit is still low compared with leaves (for ‘Hedelfinger’ fruit, up to 1.71 mm−2; Peschel et al., 2003). Attempts using SEM and EDX that allow stomata to be localized more easily were not entirely conclusive in identifying preferential precipitation of Fe in the region of the stomatal pore (M.J. Bukovac, W. Schröder, and M. Knoche, unpublished data), possibly because of signal attenuation/absorption by the cuticle. Nevertheless, we would expect ferric precipitates to form at the stomatal apparatus in an identical manner as demonstrated here for the region between stomata, thereby accounting for decreased water uptake and fruit cracking (Weichert et al., 2004; Weichert and Knoche, 2006b).
It should be noted that the presence of these precipitates in the region between stomata and the absence of effects of FeCl3 on water penetration through the CM between the stomata reported earlier (Weichert and Knoche, 2006b) are not contradictory. The data from this and our earlier study (Weichert and Knoche, 2006b) indicate that the ferric precipitates form an additional resistor arranged in series to the polar pathway or the CM. According to Ohm's law, total resistance equals the sum of individual resistances when arranged in series. From these considerations, a larger effect of ferric precipitates is expected for penetration along high-flux pathways such as the polar pathway, but a small and possibly non-detectable effect on high-resistance pathways such as the cuticle (Weichert and Knoche, 2006b). This observation would also explain why FeCl3 did not decrease transpiration of sweet cherry fruit (Beyer et al., 2002).
In summary, the data of this and our earlier studies established that under laboratory conditions, water uptake and cracking of sweet cherry fruit can be effectively controlled by a pH-dependent precipitation reaction in the fruit exocarp where the precipitates block high-flux pathways for water uptake. In contrast to non-specific coating strategies, this reaction has no side effect on transport of water vapor and most likely other gases. Unfortunately, spray application of the salts investigated so far is prohibitive for ecotoxicological reasons. However, because the mechanistic basis underlying this effect is now identified, the search for the effective and acceptable ‘silver bullet’ can now begin.
Literature Cited
Beyer, M. , Lau, S. & Knoche, M. 2005 Studies on water transport through the sweet cherry fruit surface: IX. Comparing permeability in water uptake and transpiration Planta 220 474 485
Beyer, M. , Peschel, S. , Weichert, H. & Knoche, M. 2002 Studies on water transport through the sweet cherry fruit surface: VII. Fe3+ and Al3+ reduce conductance for water uptake J. Agr. Food Chem. 250 7600 7608
Bukovac, M.J. & Petracek, P.D. 1993 Characterizing pesticide and surfactant penetration with isolated plant cuticles Pestic. Sci. 37 179 194
Franke, W. 1964 Role of guard cells in foliar absorption Nature 202 1236 1237
Furia, T.E. 1972 Sequestrants in foods 271 294 Furia T.E. CRC handbook of food additives Vol. 1 CRC Press Boca Raton, FL
Greenwood, N.N. & Earnshaw, A. 1990 Chemie der Elemente VCH-Verlagsgesellschaft Weinheim, Germany
Jones, S.S. & Long, F.A. 1952 Complex ions from iron and ethylendiaminetetraacetate: General properties and radioactive exchange J. Phys. Chem. 56 25 33
Marschner, H. 1995 Mineral nutrition of higher plants Academic Press London
Palade, G.E. 1952 A study of fixation for electron microscopy J. Exp. Med. 95 285 298
Peschel, S. , Beyer, M. & Knoche, M. 2003 Surface characteristics of sweet cherry fruit: Stomata number, distribution, functionality and surface wetting Scientia Hort. 97 265 278
Schlegel, T.K. , Schönherr, J. & Schreiber, L. 2005 Size selectivity of aqueous pores in stomatous cuticles of Vicia faba leaves Planta 221 648 655
Schönherr, J. 2006 Characterization of aqueous pores in plant cuticles and permeation of ionic solutes J. Expt. Bot. 57 2471 2491
Schönherr, J. & Schmidt, H.W. 1979 Water permeability of plant cuticles: Dependence of permeability coefficients of cuticular transpiration on vapor pressure saturation deficit Planta 144 391 400
Schönherr, J. & Bukovac, M.J. 1970 Preferential polar pathways in the cuticle and their relationship to ectodesmata Planta 92 189 201
Schönherr, J. , Fernandez, V. & Schreiber, L. 2005 Rates of cuticular penetration of chelated FeIII: Role of humidity, concentration, adjuvants, temperature, and type of chelate J. Agr. Food Chem. 53 4484 4492
Weichert, H. & Knoche, M. 2006a Studies on water transport through the sweet cherry fruit surface: X. Evidence for polar pathways across the exocarp J. Agr. Food Chem. 54 3951 3958
Weichert, H. & Knoche, M. 2006b Studies on water transport through the sweet cherry fruit surface: 11. FeCl3 decreases water permeability of polar pathways J. Agr. Food Chem. 54 6294 6302
Weichert, H. , v. Jagemann, C. , Peschel, S. , Knoche, M. , Neumann, D. & Erfurth, W. 2004 Studies on water transport through the sweet cherry fruit surface: VIII. Effect of selected cations on water uptake and fruit cracking J. Amer. Soc. Hort. Sci. 129 781 788
